Autism is a condition of early childhood onset marked by social delay, communication delay and repetitive behavior affecting 2 per 1,000 children (Fombonne, 2003). If the wider spectrum is included (PDD-NOS and Asperger's disorder), the prevalence is 67 per 1000 children (Autism and Developmental Disorders Monitoring Network, 2007). Given the severity and the prevalence, pervasive developmental disorders (PDDs) have enormous public health impact. There is general agreement that intervention in the preschool years can reduce the long-term disability of PDD, but the required components and optimal intensity of these programs remain unclear (National Research Council, 2001). In addition, recommended behavioral and educational treatments for young children with PDD are not available in all communities (Lord et al., 2005). Moreover, preschool interventions often focus on the child (Smith, 1999) leaving parents to manage the child's behavior on their own (Jacobson & Mulick, 2000). Thus, although the role of parents as partners in the treatment team is universally acknowledged, few studies have tested parent-based interventions and large-scale interventions are non-existent (Smith et al., 2007). In response to PA 07-092 (Collaborative R01s for Mental Disorders) and PA 07-085 (Research on Autism and Autism Spectrum Disorders), this application describes a five-site study to compare parent training to psychoeducation in 180 preschool-age children with PDD. Eligible subjects will be randomly assigned to receive either Parent Training (PT) or Psychoeducational Intervention (PEP) over 6 months. The primary outcomes are the change in the child's irritable and noncompliant behavior rated by the parent. Overall improvement and gains in adaptive functioning (social, communication and daily living skill domains) will be assessed by a rater who is blind to treatment assignment. Parent-child interaction will be assessed in a semi-structured lab sequence by blinded raters. The effects of PT on parental stress and the possible moderating effects of concurrent interventions will also be explored. Training and quality assurance services will be shared across sites; Yale University will provide data management and statistical services. This application is congruent with recommendations of an NIMH advisory council (Research in Psychosocial and Behavioral Interventions in Autism), which called for large-scale, randomized trials to evaluate the efficacy and promote the dissemination of psychosocial interventions in children with PDD (Lord et al., 2005; Smith et al., 2007). To conduct this study, we have assembled a consortium of experienced investigators from leading autism research programs including Yale, University of Rochester, University of Pittsburgh, Ohio State University, and Indiana University. These centers have extensive experience with conducting multisite trials (e.g., Research Units on Pediatric Psychopharmacology (RUPP) Autism Network) and with parenting training for children with autism.